SIMULATION OF PROCESS CHARACTERISTICS IN BIOGAS PLANTS BASED ON ANALYSIS OF INTERVAL DATA

Abstract

The problems of the research presented in the work relate to mathematical modeling in order to reflect the relationship between the main characteristic of the process and the factors that affect it, as well as the dynamics of the main characteristic of the process, which is determined by the acidity of the substrate in the bioreactor. To build mathematical models of both types, it is proposed to use the methods of parametric and structural identification of models of static objects and discrete models of object dynamics based on the analysis of interval data.

An universal approach based on metaheuristic optimization algorithms is proposed and substantiated for the identification of both types of models. These methods, in turn, use mechanisms of self-organization and self-adaptation in the process of finding an optimal or quasi-optimal solution. In particular, the work uses computational algorithms built on the basis of artificial bee colony algorithms. The method is implemented using data presented in interval form. The proposed universal method was tested on the construction of a mathematical model that reflects the dependence between the pH of the fermentation medium and the volume of the loaded bio-raw material in the form of its dry and liquid parts, the temperature in the bioreactor and the humidity of the dry part of the bio-raw material. Another mathematical model built in the work reflects the dynamics of the acidity indicator pH of the fermentation medium depending on the ratio of the mass of the loaded dry bio-raw material to the volume of the loaded liquid bio-raw material Both obtained interval mathematical models can be applied to control processes in biogas plants.